This article reviews recent advancements in intelligent robotic ultrasound (US) imaging systems, focusing on the integration of machine learning techniques to enhance the intelligence of robotic sonographers. The authors begin by discussing the commonly used robotic mechanisms and control techniques in robotic US imaging, highlighting their clinical applications. They then delve into the deployment of machine learning techniques, categorizing methods for achieving autonomous action reasoning into two sets: those relying on implicit environmental data interpretation and those using explicit interpretation. The article addresses practical challenges such as the scarcity of medical data, the need for a deeper understanding of physical aspects, and effective data representation approaches. Finally, it concludes by discussing open problems and potential perspectives for future research in this field. Key topics include semantic segmentation, registration, reinforcement learning, learning from demonstrations, and the integration of ultrasound physics into deep learning models. The article also emphasizes the importance of addressing ethical and regulatory issues, as well as the potential of emerging ultrasound imaging systems like optical US and soft ultrasonic patches.This article reviews recent advancements in intelligent robotic ultrasound (US) imaging systems, focusing on the integration of machine learning techniques to enhance the intelligence of robotic sonographers. The authors begin by discussing the commonly used robotic mechanisms and control techniques in robotic US imaging, highlighting their clinical applications. They then delve into the deployment of machine learning techniques, categorizing methods for achieving autonomous action reasoning into two sets: those relying on implicit environmental data interpretation and those using explicit interpretation. The article addresses practical challenges such as the scarcity of medical data, the need for a deeper understanding of physical aspects, and effective data representation approaches. Finally, it concludes by discussing open problems and potential perspectives for future research in this field. Key topics include semantic segmentation, registration, reinforcement learning, learning from demonstrations, and the integration of ultrasound physics into deep learning models. The article also emphasizes the importance of addressing ethical and regulatory issues, as well as the potential of emerging ultrasound imaging systems like optical US and soft ultrasonic patches.